OS Images |
os_Image_1(1)
|
Hardware Description |
hw_1
|
Number of Systems |
1
|
SW Environment |
non-virtual
|
Tuning |
- Workload Profile=High Performance Compute(HPC)
- Thermal Configuration=Enhanced Cooling
- Determinism Control=Manual
- Performance Determinism=Power Deterministic
- Memory Patrol Scrubbing=Disabled
- Numa Memory Domains Per Socket(NPS)=Two Memory Domains Per Socket
- L1 Stream HW Prefetcher=Enabled
- L2 Stream HW Prefetcher=Enabled
- Minimum Processor Idle Power Core C-State=No C-states
- Package Power Limit=390
|
Notes |
None
|
|
JVM Instances |
jvm_Backend_1(10)
|
OS Image Description |
os_1
|
Tuning |
- cpupower -c all frequency-set -g performance
- tuned-adm profile throughput-performance
- ulimit -n 1024000
- echo 960000 > /proc/sys/kernel/sched_rt_runtime_us
- echo 999000000 > /proc/sys/kernel/sched_latency_ns
- echo 40000 > /proc/sys/kernel/sched_migration_cost_ns
- echo 999000000 > /proc/sys/kernel/sched_min_granularity_ns
- echo 999000000 > /proc/sys/kernel/sched_wakeup_granularity_ns
- echo 9000 > /proc/sys/kernel/sched_nr_migrate
- echo 10000 > /proc/sys/vm/dirty_expire_centisecs
- echo 1500 > /proc/sys/vm/dirty_writeback_centisecs
- echo 40 > /proc/sys/vm/dirty_ratio
- echo 10 > /proc/sys/vm/dirty_background_ratio
- echo 10 > /proc/sys/vm/swappiness
- echo 0 > /proc/sys/kernel/numa_balancing
- echo 0 > /proc/sys/vm/numa_stat
- echo always > /sys/kernel/mm/transparent_hugepage/enabled
- echo always > /sys/kernel/mm/transparent_hugepage/defrag
|
Notes |
None
|
Parts of Benchmark |
Backend
|
JVM Instance Description |
jvm_1
|
Command Line |
-Xms31g -Xmx31g -Xmn28g -XX:AllocatePrefetchInstr=2 -XX:+UseParallelGC -XX:ParallelGCThreads=32 -XX:LargePageSizeInBytes=2m -XX:-UseAdaptiveSizePolicy -XX:+AlwaysPreTouch -XX:+UseLargePages -XX:SurvivorRatio=12 -XX:TargetSurvivorRatio=95 -XX:MaxTenuringThreshold=15 -XX:InlineSmallCode=11k -XX:MaxGCPauseMillis=100 -XX:LoopUnrollLimit=200 -XX:+UseTransparentHugePages -XX:TLABAllocationWeight=2 -XX:ThreadStackSize=140 -XX:CompileThresholdScaling=120 -XX:CICompilerCount=4 -XX:AutoBoxCacheMax=32 -XX:OnStackReplacePercentage=100 -XX:TLABSize=1m -XX:MinTLABSize=1m -XX:-ResizeTLAB -XX:TLABWasteTargetPercent=1 -XX:TLABWasteIncrement=1 -XX:YoungPLABSize=1m -XX:OldPLABSize=1m
|
Tuning |
Used numactl to affinitize each Backend JVM to 16 Core / 32 Threads - Group1: numactl --physcpubind=0-15,160-175 --localalloc
- Group2: numactl --physcpubind=16-31,176-191 --localalloc
- Group3: numactl --physcpubind=32-47,192-207 --localalloc
- Group4: numactl --physcpubind=48-63,208-223 --localalloc
- Group5: numactl --physcpubind=64-79,224-239 --localalloc
- Group6: numactl --physcpubind=80-95,240-255 --localalloc
- Group7: numactl --physcpubind=96-111,256-271 --localalloc
- Group8: numactl --physcpubind=112-127,272-287 --localalloc
- Group9: numactl --physcpubind=128-143,288-303 --localalloc
- Group10: numactl --physcpubind=144-159,304-319 --localalloc
|
Notes |
None
|
|